This paper examines the impact of India's vast railroad network on the country's economy, using archival data from colonial India. The author, Dave Donaldson, employs a Ricardian trade model with multiple regions and commodities to estimate the effects of railroads on trade costs, interregional price gaps, trade flows, and real income levels. The model predicts that railroads should reduce trade costs, increase trade, and boost real income, with a sufficient statistic for the welfare gains from railroads being the share of a district's expenditure sourced from itself. Empirically, Donaldson finds that railroads significantly reduced trade costs, increased trade flows, and raised real income by 16%. The model also suggests that about half of the welfare gains from railroads can be attributed to reduced trade costs. The paper contributes to the literature on the economic effects of large infrastructure projects and the welfare effects of trade openness.This paper examines the impact of India's vast railroad network on the country's economy, using archival data from colonial India. The author, Dave Donaldson, employs a Ricardian trade model with multiple regions and commodities to estimate the effects of railroads on trade costs, interregional price gaps, trade flows, and real income levels. The model predicts that railroads should reduce trade costs, increase trade, and boost real income, with a sufficient statistic for the welfare gains from railroads being the share of a district's expenditure sourced from itself. Empirically, Donaldson finds that railroads significantly reduced trade costs, increased trade flows, and raised real income by 16%. The model also suggests that about half of the welfare gains from railroads can be attributed to reduced trade costs. The paper contributes to the literature on the economic effects of large infrastructure projects and the welfare effects of trade openness.